Validating Machine Learning Approaches for Predicting Drug Sensitivities in Cancer Cell Lines from Functional Genetic Screens
Open Access DepositedFunctional genetic screens are widely utilized for repurposing existing drugs and identifying potential new drug targets. However, understanding the precise associations between hits in these screens and candidate drugs remains a challenge. In this study, we present a systematic approach to predict drug responses from functional genetic screens in an unbiased manner. Leveraging DepMap data, which includes extensive CRISPR-Cas9 screens and the growth inhibitory activities of over 4,000 compounds, we developed a predictive pipeline using various machine learning methods such as random forest, Support Vector Machine (SVM), Gradient Boosting Decision Trees (GBDT), and XGBoost. Our classification algorithms demonstrated robust predictive power, revealing numerous known associations between gene perturbations and drug responses. Our established pipeline serves as a predictive tool capable of forecasting drug sensitivities in cancer cell lines. The significance of this work lies in its potential to discover novel therapeutic strategies and explore opportunities for drug repurposing. By offering an in-silico approach to drug screening derived from functional genetic screens, our study contributes to accelerating drug discovery and facilitating the identification of promising avenues for cancer treatment.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.